Blocks not being easily verifiable is a quite a bum. Some part of the community is trying to make lightclient based zk bridges, but it’s basically makes it impossible for avalanche…
Yeah it's not ideal for supporting light clients. They have implemented BLS signatures for their Warp Messaging protocol which should be the equivalent of lightclient based bridging and should be useable for a zk bridge. I know they have this for subnets but I'm not sure about the primary chains
I’d love to see a side by side comparison video of Avalanche vs Ouroboros consensus models and see the pros and cons summarized at the end. And compare the scalability showing how they make subnets and cardano sidechains
Essentially Avalanche is much much faster whereas Ouroboros is one of the slowest consensus methods outside of bitcoin. The advantage of Ouroboros is that it will stay online even if the majority of participants go offline for whatever reason whereas Avalanche is set up to freeze if only 20% of participants are offline. Avalanche is designed for speed whereas ouroboros is designed for robustness. I may make a video on subnets in the future but I personally don't think subnets or side chains are particularly good for scaling and rollups/validiums are the far superior technology. Subnets make the assumption that a subset of the stakers are honest but you can only make this assumption if you are randomly selecting the subset. If you are hand selecting the stakers like subnets then you can hand pick a subset of malicious stakers to run the subnet and eventually steal everyone's funds. I personally don't think they are that different to sidechains and don't think either are good long term solutions.
@@altexplainerok interesting. I’ve been very interested in learning about how these Avalanche and Ouroboros as well as Polkadots consensuses allow anyone to make their own custom brand new blockchain and plug into the consensus model each has and attain “automatic decentralization”. The reason to connect all these blockchains is to improve interoperability and communication between blockchains which they say is absolutely going to be necessary.
You can't attain "automatic decentralisation" from just using a certain consensus model. The decentralisation part comes from how decentralised the participants of the consensus model are. If you only have 1 person with 100 computers participating in the consensus model, it's not decentralised. Interoperability is not really possible in the way people talk about it. The bitcoin token is only the bitcoin token when it is secured by bitcoin nodes. If there is a bitcoin token being secured by polkadot nodes, it's no longer a bitcoin token as the security properties are different. It is now a wrapped bitcoin token. You can't actually transfer assets between blockchains without changing the security properties. Some methods like sharding and rollups allow for almost the same security properties between chains and so for example people don't refer to ETH on a rollup chain as wrapped ETH but instead as just ETH as the security is almost the same as ETH on the L1 but really it is also wrapped ETH under the hood. If you don't care about maintaining security when transferring between chains then that's not a difficult problem and was solved already with any bridge. When you send BTC to Coinbase that means bitcoin is interoperable with Coinbase. For me, interoperability is just a marketing term and is either impossible or already possible and very easy.
@@altexplainer what do you think about how in Ouroboros proof of stake, the person that wins the slot doesn’t have to be limited to only making 1 block for Cardano, but also has the ability to make 2 or 3 or more blocks at the same time while they’re in that slot, and the notion that several of these stake pool operators can make several blocks at once while they’re in a slot in an epoch? Like what may happen when Midnight and other chains join the network. And they’re incentivized to do multiple blocks for multiple chains to get rewards for the blocks created by those chains as well. And delegators will also want to delegate to pools that are giving them multiple rewards on multiple chains. So when somebody wants to build a brand new blockchain, rather than having to figure out their own science to a consensus mechanism, and then recruit people from scratch to join their network, the can join an already existing network like Cardano, Avalanche, Polkadot, or whichever they like. They have that option too.
If the validators are creating more blocks that means more work for nodes which leads to node requirements increasing. Increasing the node requirements means less people can afford to run a node at home meaning the network is less decentralised. If you are going the route of making validators require powerful hardware, Solana is already very good at this. Rollups/validiums are the answer to someone wanting to build a new chain without having to build a consensus and recruit people from scratch. There are teams like sovereign labs who I think are working on making it easy for anyone to launch a rollup. Imo, if the scalability solution is not using zero knowledge proofs, it's an outdated design and rollups are probably better.
Love to see how AVAX consensus compares to Hashgraph ( Hedera). It's kinda interesting that the criticism of hashgraph (on wikipedia en.wikipedia.org/wiki/Hashgraph) is attributed to Emin, one of the founders of Avalanche.
![DDG & G Herbo - Nosey [Official Video]](http://i.ytimg.com/vi/I9Ra0Zc2NlQ/mqdefault.jpg)
This is the best explanation for Avalanche I’ve found
Great Video
Blocks not being easily verifiable is a quite a bum. Some part of the community is trying to make lightclient based zk bridges, but it’s basically makes it impossible for avalanche…
Yeah it's not ideal for supporting light clients. They have implemented BLS signatures for their Warp Messaging protocol which should be the equivalent of lightclient based bridging and should be useable for a zk bridge. I know they have this for subnets but I'm not sure about the primary chains
Can you make a video on DJED and SHEN?
I'll add them to my list of video idea's but I haven't read that much about stablecoins so will probably be a low priority for now
@@altexplainer Thanks, particularly how DJED compares to indigo iUSD
I’d love to see a side by side comparison video of Avalanche vs Ouroboros consensus models and see the pros and cons summarized at the end. And compare the scalability showing how they make subnets and cardano sidechains
Essentially Avalanche is much much faster whereas Ouroboros is one of the slowest consensus methods outside of bitcoin. The advantage of Ouroboros is that it will stay online even if the majority of participants go offline for whatever reason whereas Avalanche is set up to freeze if only 20% of participants are offline.
Avalanche is designed for speed whereas ouroboros is designed for robustness.
I may make a video on subnets in the future but I personally don't think subnets or side chains are particularly good for scaling and rollups/validiums are the far superior technology.
Subnets make the assumption that a subset of the stakers are honest but you can only make this assumption if you are randomly selecting the subset. If you are hand selecting the stakers like subnets then you can hand pick a subset of malicious stakers to run the subnet and eventually steal everyone's funds. I personally don't think they are that different to sidechains and don't think either are good long term solutions.
@@altexplainerok interesting. I’ve been very interested in learning about how these Avalanche and Ouroboros as well as Polkadots consensuses allow anyone to make their own custom brand new blockchain and plug into the consensus model each has and attain “automatic decentralization”. The reason to connect all these blockchains is to improve interoperability and communication between blockchains which they say is absolutely going to be necessary.
You can't attain "automatic decentralisation" from just using a certain consensus model. The decentralisation part comes from how decentralised the participants of the consensus model are. If you only have 1 person with 100 computers participating in the consensus model, it's not decentralised.
Interoperability is not really possible in the way people talk about it. The bitcoin token is only the bitcoin token when it is secured by bitcoin nodes. If there is a bitcoin token being secured by polkadot nodes, it's no longer a bitcoin token as the security properties are different. It is now a wrapped bitcoin token. You can't actually transfer assets between blockchains without changing the security properties.
Some methods like sharding and rollups allow for almost the same security properties between chains and so for example people don't refer to ETH on a rollup chain as wrapped ETH but instead as just ETH as the security is almost the same as ETH on the L1 but really it is also wrapped ETH under the hood.
If you don't care about maintaining security when transferring between chains then that's not a difficult problem and was solved already with any bridge. When you send BTC to Coinbase that means bitcoin is interoperable with Coinbase.
For me, interoperability is just a marketing term and is either impossible or already possible and very easy.
@@altexplainer what do you think about how in Ouroboros proof of stake, the person that wins the slot doesn’t have to be limited to only making 1 block for Cardano, but also has the ability to make 2 or 3 or more blocks at the same time while they’re in that slot, and the notion that several of these stake pool operators can make several blocks at once while they’re in a slot in an epoch? Like what may happen when Midnight and other chains join the network. And they’re incentivized to do multiple blocks for multiple chains to get rewards for the blocks created by those chains as well. And delegators will also want to delegate to pools that are giving them multiple rewards on multiple chains.
So when somebody wants to build a brand new blockchain, rather than having to figure out their own science to a consensus mechanism, and then recruit people from scratch to join their network, the can join an already existing network like Cardano, Avalanche, Polkadot, or whichever they like. They have that option too.
If the validators are creating more blocks that means more work for nodes which leads to node requirements increasing. Increasing the node requirements means less people can afford to run a node at home meaning the network is less decentralised. If you are going the route of making validators require powerful hardware, Solana is already very good at this.
Rollups/validiums are the answer to someone wanting to build a new chain without having to build a consensus and recruit people from scratch. There are teams like sovereign labs who I think are working on making it easy for anyone to launch a rollup.
Imo, if the scalability solution is not using zero knowledge proofs, it's an outdated design and rollups are probably better.
I love all the videos you provide
Great content... subbed
Simple, concise and understandable!!
Love to see how AVAX consensus compares to Hashgraph ( Hedera). It's kinda interesting that the criticism of hashgraph (on wikipedia en.wikipedia.org/wiki/Hashgraph) is attributed to Emin, one of the founders of Avalanche.
I plan on covering Hedera eventually but right now I'm not that familiar with it so it'll probably be a while